Management final report about survey
MGMT203 Data Analysis and Research in Business
Fall 2016
PROJECT GUIDELINES AND EXPECTATIONS
Contents1
1
1. Project Goals 2
2. Team size 2
3. Project Idea Proposal 2
4. Software Use 3
5. Survey Design (Progress Report) 3
6. Oral Presentation 4
7. Final Report 4
8. Scale Design for Questionnaires 9
9. Sample Table Formats 11
10. Frequently made mistakes 14
11. Grading Templates 15
1. Project Goals
The purpose of the group project is to help you become an intelligent consumer of statistics, by helping you learn how to do a “real-world” statistical analysis including the process of coming up with a relevant research question, obtaining and working with data, conducting descriptive analysis, and drawing conclusions. Furthermore, you need to understand the downsides of your analysis and what you could do if you were to pursue this project in the future.
The quality of your work and the amount of effort you put in towards your project will be very important in your overall project grade. Expectations shown below apply to the entire project.
2. Team sizeThe project is to be done in groups of maximum three, freely formed among the members of the class. No individual projects will be allowed, so please be sure to find two partners and to communicate the members of the group before the deadline on course syllabus.
3. Project Idea ProposalYou will be required to write a project idea (2-3 paragraphs, 1.5 spaced in 12 pt text 200-300 words) and submit it by the deadline on the syllabus. The proposal must be done for 5 to 10 variables, in addition to any demographic questions you ask. A variable could be associated with one or more questions in the survey.
You may propose any relevant project related to Business/Management, but it should not be about measuring general satisfaction with any service on campus (e.g. dormitory, shuttles). Each group must come up with a project from a certain point of view, with a clear objective and must propose solutions as a result of their analysis. Example topics could be such as:
A hotel manager would like to better understand the ratings of their customers. When customers stay, upon their leave they fill out a questionnaire. Assume that your group has such dataset with customer ratings. Your group objective will be to assist the hotel manager in better understanding the customer reviews by using data analysis tools you have learnt during the course. The output from this project may involve suggestions for the hotel manager (for example: increase cleaning services, pay better attention to security etc.).
A company wants to launch a new air conditioner. They have collected data through a marketing research company. The objective for the group will be to analyze the data in helping the company make the right decision on the specifying air condition features, i.e. low noise, high power, medium cost etc.
You should have some specific hypothesis that you would like to test regarding the topic, and you should write them in your proposal. For example, for the hotel manager’s problem above, you may want to test the hypothesis whether foreign tourists and Turkish tourists have different satisfaction levels regarding their rooms. Or for air conditioner company, you may want to test the hypothesis that people who want low noise are willing to pay for that feature. Note that such hypothesis would have an effect on the questions you ask and the type of scale you use for these questions.
When explaining the purpose and scope of your project you must explicitly state what your target population is. This is also very important because you will need to reach a sample that represents this target population.
For your project to be accepted it must be:
(1) Relevant – you must demonstrate why your project question is important (to you, school, your work, or world in general).
(2) Doable within the given time frame – it must be evident that you will be able to obtain necessary data in time and be able to use the obtained data to perform analysis that addresses your project question.
(3) One that does not address the same questions as someone else’s project (past or current).
Do not forget to write all team members’ names on the proposal document.
4. Software UseIf you plan to have an online survey, then you must use Google forms. No other tool will be accepted. When you submit your progress report, you could submit a link to the Google forms survey you have designed.
5. Survey Design (Progress Report)In designing the survey, students should be using the survey design guidelines provided both in this document and the lecture slides/notes. At least three questions should use a five-point scale (such as the Likert scale). You should not re-invent the wheel when designing surveys. Most of the concepts you want to test or research have been done so in the past. So researching literature, checking text books is a good starting point. For example, service quality is an important research area and there are textbooks (on Quality Management), standard questionnaires (e.g. SERVQUAL) that you can make use of.
If you have concepts or constructs such as reliability, loyalty you should start of by picking a definition from the literature. People may have different understanding of such constructs and unless you start with a definition, the questions you come up with may not be accurate. Making use of the relevant academic literature would help you in this respect.
6. Oral PresentationYou will present your project with a 10 minute-presentation on an appointed time and date. You can find the grading template with evaluation criteria at the end of this document.
7. Final ReportThe “Table of Contents” of the report should include the following (the full report would be around 10-15 pages not including cover, table of contents, summary and appendices; 2000-3000 words; 1.5 spaced in 12 pt text with 2.5 cm. margins):
Executive Summary
Overview and Background of Situation
Detailed Statement of Problem
Research Design
Data Analysis
Data Pre-processing
Description of Sample Characteristics
Data Analysis
Conclusions
Appendices (as per need)
The report should not require the reader to read or check the Excel files. All important comments, figures etc. that are from the Excel files should be included in the report in a proper format (which may be different from the format in the Excel file).
Figures and tables should be numbered (Figure 1, Figure 2, Table 1, Table 2, etc.) and they should have appropriate captions. In the text, you should refer to the figures and tables using these numbers.
Pages of the report should be numbered, the table of contents should refer to these page numbers as well.
You should use MS Word’s spell-checker to make sure there are not spelling or grammar mistakes in your report.
You can find the grading template with evaluation criteria at the end of this document.
Below you can find explanation of what you should provide under each section of the report.
Cover page: Your cover page should be prepared using the following format.
Table of Contents: Should have page numbers. Microsoft Word has an automatic feature to add a table of contents with the page number. You MUST learn and utilize this (it is under the References tab of the Ribbon). Note that, for MS Word to create a table of contents the titles of the sections must be formatted by the default heading styles (see Styles group on the Home tab of the Ribbon).
1. Executive Summary: At most one page summary of the project and your findings. This is the final piece that you should write in your report. It must motivate the reader (a manager with very limited time) to read the rest of the report. You should specify date (or dates) of data collection, sample size, sample selection method, types (categories) of questions asked and main purposes (objectives) of the project. Key findings and conclusions also be given.
2. Overview and Background of Situation: This section must have two parts:
The case: Give brief background information about the case. Try to relate it to “the detailed statement of problem” (Section 3) on hand. Note that this could be a hypothetical case. In that case, you will have to list your assumptions.
Summary of the problem: give the summary of the business problem in two or three sentences. Detailed problem statement will be written in the next section.
3. Detailed Statement of Problem: Here you will give the details of the problem. Note that the problem must be business/management related. A bad example would be: “The ABC Company wants to analyze the data”. The problem must be addressing a specific business problem such as: “The marketing department has been asked to launch a new environmentally friendly shampoo. They would like to better understand their consumer segments…”. The problem statement must consist of two parts:
3.1. Managerial/Business Problem: Here, elaborate on the problem on hand, from managerial point of view.
3.2. Business Data Analysis Research Problem: Here, establish clear objectives for the research project in order to solve the managerial/business problem. This means you should state:
What are the metrics, variables, data that is need to solve the managerial problem,
What specific questions (e.g. “Does gender affect students’ satisfaction with the cleaning services at the dormitories”) you will try to answer using the data. These are typically a set of hypothesis that you would like to test.
4. Research Design: The following subtitles must be used to explain the research design.
Description of Data Collection Technique or Data Source (When, Where, and How)
Questionnaire questions (What): All questions and their answer choices as they appeared in the survey.
Meta-Data structure (variable names, variables description, missing value count...). (e.g., variable name “CLEAN” represents the answers given to the questions “How satisfied are you with the cleanliness of the dorms?”, and “10 of the 120 respondents did not answer this question).
Clear description of the target population
Sampling frame, sample size and sampling method(s), response rate
A brief discussion of why the resultant sample is representative of the target population
5. Results and Data Analysis: A typical project must address the following issues:
Data pre-processing: proper formatting in Excel, making sure that the data contains no error. Note that the final dataset used will be attached as a separate Excel file to the project report:
Outlier detection (if applicable): Is outlier detection meaningful? Should we delete outliers? Are there obvious outliers? Why did this happen? Did you use any outlier elimination technique, such as Winsorising the dataset?
Missing Value Analysis (if applicable): Are there questions that were not answered by respondents? Are there empty cells in your dataset? Is there a variable with too many missing values? Which of the following method is the most appropriate when tackling the issue of missing values? (Such as: Imputing by average, by mode, or by median)
Description of Sample or Dataset Characteristics
Individual question level results (descriptive statistics such as average etc.) and any appropriate charts (such as pie chart etc.).
Relationships between answers to different questions: Cross-tabulations or contingency tables, correlations between answers to different questions, appropriate charts, etc. Remember that Excel pivot table tool can help you produce cross-tabulations.
Analysis using one or more of the following techniques
Hypothesis Testing, Regression Analysis, One-Way ANOVA, Chi-square test of independence.
6. Conclusions: 1-2 page results summary. This includes a summary of:
Any limitations of the study
Conclusion from the results (the recommended solutions)
Managerial implications and discussion
7. Appendices: (as per need, and possibly as separate document attachments. You may add other relevant documents if needed)
Survey (full text)
Dichotomous Scales
A dichotomous scale is a two-point scale which presents options that are absolutely opposite each other. This type of response scale does not give the respondent an opportunity to be neutral on his answer in a question.
Examples:
Yes- No
True - False
Fair - Unfair
Agree – Disagree
Rating Scales
Three-point, five-point, and seven-point scales are all included in the umbrella term “rating scale”. A rating scale provides more than two options, in which the respondent can answer in neutrality over a question being asked.
Examples:
1. Three-point Scales
Good - Fair – Poor
Agree – Undecided - Disagree
Extremely- Moderately - Not at all
Too much - About right - Too little
2. Five-point Scales (e.g. Likert Scale)
Strongly Agree – Agree – Undecided / Neutral - Disagree - Strongly Disagree
Extremely – Very - Moderately – Slightly - Not at all
Excellent - Above Average – Average - Below Average - Very Poor
3. Seven-point Scales
Exceptional – Excellent – Very Good – Good – Fair – Poor – Very Poor
Very satisfied - Moderately satisfied - Slightly satisfied – Neutral - Slightly dissatisfied - Moderately Dissatisfied- Very dissatisfied
Acceptability | Not at all acceptable, Slightly acceptable, Moderately acceptable, Very acceptable, Completely acceptable |
Agreement – 1 | Completely disagree, Disagree, Somewhat disagree, Neither agree nor disagree, Somewhat agree, Agree, Completely agree |
Agreement – 2 | ompletely True, Somewhat True, Slightly True, Neither True nor False, Slightly False, Somewhat False, Completely False |
Appropriateness | Absolutely inappropriate, Inappropriate, Slightly inappropriate, Neutral, Slightly appropriate, Appropriate, Absolutely appropriate |
Awareness | Not at all aware, Slightly aware, Moderately aware, Very aware, Extremely aware |
Beliefs | Not at all true of what I believe, Slightly true of what I believe, Moderately true of what I believe, Very true of what I believe, Completely true of what I believe |
Concern | Not at all concerned, Slightly concerned, Moderately concerned, Very concerned, Extremely concerned |
Familiarity | Not at all familiar, Slightly familiar, Moderately familiar, Very familiar, Extremely familiar |
Frequency | Never, Rarely, Sometimes, Often, Always |
Importance | Not at all important, Slightly important, Moderately important, Very important, Extremely important |
Influence | Not at all influential, Slightly influential, Moderately influential, Very influential, Extremely influential |
Likelihood | Not at all likely, Slightly likely, Moderately likely, Very likely, Completely likely |
Priority | Not a priority, Low priority, Medium priority, High priority, Essential |
Probability | Not at all probable, Slightly probable, Moderately probable, Very probable, Completely probable |
Quality | Very poor, Poor, Fair, Good, Excellent |
Reflect Me | Not at all true of me, Slightly true of me, Moderately true of me, Very true of me, Completely true of me |
Satisfaction (bipolar) | Completely dissatisfied, Mostly dissatisfied, Somewhat dissatisfied, Neither satisfied or dissatisfied, Somewhat satisfied, Mostly satisfied, Completely satisfied |
Satisfaction (unipolar) | Not at all satisfied, Slightly satisfied, Moderately satisfied, Very satisfied, Completely satisfied |
When respondents chose the midpoint (“Neither agree nor disagree”), it is generally a valid response (Narayan & Krosnick, 1996; O’Muurcheartaigh, Krosnick & Helic, 1999)
In agreement scales, the element of “strongly” can confound the emotional strength component with the cognitive agreement task (Fowler, 1995)
Bipolar scales (Disagree to Agree) have a maximum reliability and validity at 7 points whereas unipolar scales (e.g., Not True at all to Completely True) have a maximum reliability and validity at 5 points (Krosnick & Fabrigar, 2003)
Numeric labels seem to increase confusion rather than verbal labels (Krosnick & Fabrigar, 2003)
Agree/Disagree scales are less desirable than True/False scales for these reasons (Fowler, 1995):
Agree/Disagree questions tend to be cognitively complex. For example, disagreeing that one is seldom overwhelmed by life stressors is a complicated way of saying that one is often overwhelmed
Research has consistently demonstrated a tendency of less educated respondents toward acquiescence, which leads them to be more likely to “agree” categories
In particular, using “strongly” agree/disagree actually violates a question design mode because it contains 2 dimensions—an emotional strength component and cognitive agreement task
n Single item scales are typically not recommended except for measures of job satisfaction (Dolbier, Webster, McCalister, Mallon & Steinhardt, 2004) and specific behaviors (e.g., smoking, drinking) due to low internal consistency reliabilities.
(Source: Zikmund et al. Business Research Methods, 8th ed.)
Parts of a table
Reporting format for a typical cross-tabulation
Reporting format for a typical statistical test
Using a stubhead format to include several cross-tabulations in one table
The report should be self-contained – the reader should not be referred to any other document, file, web site etc. to get some information or understand the report.
Linear regression models cannot have a dependent variable with 0-1 values.
Table and charts must have numbers (such as Table 1, Table 2, Chart 1, etc.) and captions and in the main text of the report you should refer to these table and figure numbers.
Your main charts and figures must be inside the report (not left in Excel or in the appendix). The report should give all the essential information without any need to check the Excel files.
For questions that you asked using the Likert scale the Descriptive statistics of Section 5 should give averages and correlations between potentially related questions.
Spelling mistakes: It is unforgivable not to use MS Word’s spell-checker. If you have any spelling mistake that can be caught by the spell-checker you will lose 10% of the report grade that is associated with the “well-written” criterion.
Survey questions must have numbers. You should refer to these numbers when discussing the results.
With nominal data you should only report percentages, nor average and standard deviation.
With nominal data rather than reporting correlation provide a contingency table.
Oral Presentation Grading Template
Final Report Grading Template
17/17